THE IDENTIFICATION AND ANALYSIS OF HUMAN CAUSATION IN ATROCIOUS
AND ABNORMAL NATURAL DISASTERS BY THE METHOD OF SATELLITE
REMOTE SENSING
Na Yang *, Liangming Liu
School of Remote Sensing and Information Engineering, Wuhan University, Wuhan Hubei China, 430079, yangna3565@yahoo.com.cn or lm_liu69@sohu.com
Commission VIII, WG VIII/3
KEY WORDS: Disaster, Sustainable, Climate, Inspection, Modelling, Analysis
ABSTRACT:
On the basis of climate change and natural disasters in China during the past ten years, and with the help of satellite remote sensing
data (MODIS), we make experiments on the Heat Island Effect of Beijing, vegetation coverage in the Greater and Xiaoxing’anling
Mountains, and heat distribution in the Yangtze River Basin. By making use of an integrative collection of variety information
including population distribution, land coverage and utilization, super-scaled projects and so forth, we calculate Land Surface
Temperature (LST) and Normalized Difference Vegetation Index (NDVI) in order to find clues to prove that human beings have the
responsibility for causing atrocious and abnormal natural disasters.
1. INTRODUCTION
The Earth is compelled to suffer a challenge of vital importance
without any willingness. With characteristics of arbitrariness in
landform changing, immoderation in natural resources
exploitation and irrationality in large-scaled projects
construction, it seems those influences that we have imposed on
the Earth are negative and devastating in the past hundred years,
which we call times of Industrial or Science-Technology
Revolution.
Take Beijing for example, the statistical data says that its
resident population is 16.33 million in 2007; the density in the
centre of capital is 22394 person-per-km2. With MODIS data,
we make studies on the Heat Island Effect (Gao Mao-fang et al.,
2007; Yan Feng et al., 2007; Nitin K. Singh & Sreekala G.
Bajwa, 2007) of Beijing during the period of 2003~2007, and
figure 1 is the land surface temperature (290~320K, MODIS,
1km) of this area in 2003-09-10 (U_L), 2005-09-14 (U_R) and
2007-09-04 (Below) respectively, we can see clearly that the
temperature has a feature of annular distribution and the central
region with high temperature is growing annually.
The complexity of forecasting and predicting natural disasters
doesn't become easy along with our increased knowledge; on
the contrary, with their newly evolved characteristics such as
great unpredictability, irregularity and diversity, researches that
focus on their performances seem only treating the symptoms
but not the root causes. In fact, human beings have so special
position in biological chain and the Earth’s Layers, that our
status have gradually changed from victims to accomplices or
even murderers, when factors of society, politics, economy and
culture have more and more subjective authorities in the
environment decision making.
2. A FRAGILE TREND OF THE CITY ECOLOGICAL
ENVIRONMENT
Big cities (regions) with a great density of population coincide
with centres of high temperature on images of satellite remote
sensing. These dense regions with low coverage of vegetation
have extremely high release of carbon dioxide and exhaust heat;
their function of heat accumulation is so great that could make
direct influence on the regional energy cycle and trigger
phenomena like drought, rainstorm and sandstorm. On the other
hand, processes of energy transformation and diversion have
been changed from complex to simple as a result of the
simplification of biodiversity, the city ecological environment is
being driven to a frangible direction and having its capability of
resisting natural disasters lost.
Figure 1. Heat Island Effect of Being in 2003, 2005 and 2007
We draw a white frame around the city main body, and give its
temperature grades as the little picture in upper-right, the little
grey one in lower-right is the reflectance of MODIS band 2 for
the whole region.
We make four temperature grades of the city main body with
different population density, as shown in table 1: the capital-
535
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B8. Beijing 2008
core-area (Orange) grows fastest with top population density;
the function-expanded-region (Yellow) comes to heel, and the
new-developed-area (Green) seems having not much change,
while the ecological-conservation-zone (Blue) which should
have high vegetation coverage undergoes a sharp decrease.
Population
Density
22394
6312
709
200
person-perkm2
Temperature
2003
2005
2007
＞ 309
305 ~ 309
301 ~ 305
＜ 301
0.3
18.67
62.14
18.66
4.02
40.4
50.12
5.36
6.2
33.45
53.94
6.3
K
%
%
%
3. INSPECTION OF DEGENERATE PROCESSES
Satellite remote sensing has the predominance of long-time
monitoring and change detection. With years-maps of land
classification, we could get an overall trend on courses of
degradation like desertification, soil-and-water loss. The
quantificational information of woodland, farmland and
residential-land which obtained from satellite remote sensing
data reflects the transformation of the atmospheric underlying
surface (He Xuezhao et al., 2002). The land type alteration
would change or even interrupt the local energy and water cycle
in quite a short period and therefore, has un-shirkable
responsibilities in happenings of disasters like drought, flood,
rock-mud flow etc.
The forests in the Greater and Xiaoxing’anling Mountains have
key role in climate regulation of North-East China. We
calculate the mean NDVI (2000~2007, MODIS, 1km) of this
region from June to September which shown in figure 3.
Table 1. Temperature grades and their ratio in area
For verifying the changing-relationships among land types, land
surface temperature and vegetation coverage, we choose points
of 2003 and 2007 in a line of 39.92N, 115.19~117.69E. Solid
lines of red, green and dark-blue in figure 2 represent LST
(290~320K, MODIS, 1km), NDVI (0~0.8, MODIS, 1km) and
reflectance (0~0.8, MODIS, 1km) distributions of 2003, and
lines of pink, light-green and sky-blue with diamonds are for
values of 2007 (the sharp downside in 115.7E is caused by
cloud).
Figure 2. Reflectance, NDVI and LST in the line of 39.92N,
115.19~117.69E, 2003 and 2007
We can see dramatically opposed characteristics of land
coverage and development trend in city-centre and suburban
regions: the former has higher temperature and lower vegetation
coverage, and the latter is exactly reversed. The overall trend
of NDVI is decreasing from 2003 to 2007, but the temperature
has an opposite trend in this period, and a wide range appears in
its middle part (city centre). The reflectance of suburban areas
also shows a remarkable increase and this reflects the city’s
expansion and its coverage transformation to some extent.
Figure 3. Mean NDVI of Heilongjiang and north-east Inner
Mongolia, June~September, 2000~2007, -0.2~0.8
There is an obvious increase in the picture of 2001, and then
follows a sharp and successive decrease in 2002 and 2003, the
truth is that continuous drought of large-scale happens in the
spring and some places of this region even get their first rain till
mid-July that year. Take 2004 as the beginning, the NDVI
values get a rebound and exceed 2000’s level gradually.
The sharp decrease of the ecological-conservation-zone and fast
explosion of the capital-core-area from 2003 to 2007 makes
great contrast: alterations of land coverage and regional
hydrothermal conditions within a short period of time will put
heavy load to the self-adaptive cycle of energy and water,
therefore, coming along those extreme or abnormal weathers,
and some times frequently.
We choose values located in lines of 51.9846N, 48.1857N,
44.9525N respectively, and analyze their distribution in years
of 2001~2007, with the 2000’s a standard.
The y-axis in figure 4 represents the ratio that other years’
NDVI values greater than those of 2000’s and the x-axis is for
different years. There are cities located in the line of low-
536
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B8. Beijing 2008
the maximum since 1951. There we could say that all this is
only a little tip of the iceberg of the global warming.
latitude (44.9525N, Green), and its NDVI values are always
higher than that of the 2000’s, but having an overall decreasing
trend, this just accords with the fact of city expansion and its
land type-changing. The mid-latitude line (48.1857N, Blue)
which goes across the Greater xing’an Mountains and central
areas of Heilongjiang province has a continuous decrease in
2003 and 2004, but the whole level is above the 2000’s and we
can see a slightly upward trend. The high-latitude line
(51.9846N, Red) passes through forests of Xiaoxing’anling
Mountains, there is a big fluctuation from 2000~2004, and the
overall trend wanders about the line of 50% with no explicit
increase.
4. CLIMATE CHANGE CAUSED BY SUPER-SCALED
ARTIFICIAL CONSTRUCTIONS
These super-scaled artificial constructions have a far-reaching
influence on the global climate. In fact, this holistic impact gets
a distinct “Buffer” course from infliction to being noticed, and
meanwhile, an inertial character emerges and the influence is
irreversible in quite a long time when this happened.
Take the Three Gorges Dam for example, as a world-famous
water conservancy projects, its impact on the global climate is
also world-class. We suppose the gigantic intercept and
climbing water-level will inevitably remodel the heat
distribution of the Yangtze River Basin, the old pattern has
been broken and the new one is growing gradually.
We calculate the LST (mean from October to December) and
NDVI of the Yangtze River Basin from 2000 to 2007 with
MODIS data (1km). In figure 6 there are LST (273~313K) and
NDVI (-0.4~0.7) of 2000, 2003 and 2007 respectively, though
we don't have enough satellite data to prove whether those heat
centres enhanced or diverted just after the construction of this
great dam in 1993, we see clearly that real changes have
happened: banded zones of temperature that approximately
paralleling the latitude have been replaced by enhanced heat
centres. The heat accumulation in provinces of Hubei, Hunan
and Jiangxi has an enlarging trend in both area and intensity.
While as we pay attention to those NDVI pictures, there seems
no violent change happened to this area’s vegetation coverage,
there we may have the conclusion that the heat remodelling is
not simply caused by any processes of degradation or the result
of urbanisation. The Yangtze River just likes a huge channel for
heat and energy transformation, and close it down will
definitely affect the global climate sooner or later: induce a new
pattern, bring along extreme weather phenomena, or even form
entirely different climate zones.
Figure 4. Ratios of 2001~2007 compared with 2000’s in latitude
of 51.9846N_R, 48.1857N_B and 44.9525N_G
Figure 5 makes us see the fluctuation clearly in the line of
51.9846N during 2001~2004, the y-axis is for 2000 and the xaxis is for the other years: most values concentrate on the side
of 2001 at first, then completely reversed in pictures of 2002
and 2003, while in 2004, there comes a balance, just as shown
in figure 3 and 4.
Figure 5. Comparing NDVI values in the line of 51.9846N,
2001~2004, -0.2~0.8
According to the vegetation coverage, we find no trace for
processes of deforestation and desertification in the period of
2000~2007, there seems to be a stable development of the
ecological environment in this region. However, from the end
of last century, abnormal phenomenons like “dark-day in
rainstorm” and “snow in May” appear in many places of
Heilongjiang province and records for drought, high
temperature and sandstorm rise annually. Heavy snows in the
winter always come late in the early spring. In mid-April 2005,
there is such a heavy blizzard in the Songnen Plain that to be
Figure 6. LST and NDVI of the Yangtze River Basin, 2000,
2003 and 2007
537
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B8. Beijing 2008
We also make figure 9 for seeing those cities’ temperature
distribution clearly and Wuhan to be the standard. The star,
triangle and diamond represent Shanghai, Hangzhou and
Nanchang respectively, and it seems truly that new “stove”
cities have been born for most maximum values concentrate
below the diagonal. But the case of higher temperature in the
level of minimum for Wuhan, explains the influence of relativewarm water effect from the Yangtze River to some extent.
Frankly speaking, abnormal weather conditions have happened
in this region, like strong typhoons with high frequency,
successive high temperature of Chongqing in the summer of
2006, and later a warm winter for Wuhan, while in the winter of
2007, the whole Central China suffers an unprecedented snowice disaster.
We make 17 temperature grades (see appendix) and calculate
their proportions in this area, figure 7 shows the results of 2001
(Blue), 2003 (Green), 2005 (Pink) and 2007 (Red), the overall
trend is moving to higher temperature, and the difference
among grades is diminishing, grades represent low temperature
have their proportions decreasing, and increasing for those
representing high temperature.
Figure 9. LST comparison among cities of Wuhan, Nanchang
(Diamond), Shanghai (Star) and Hangzhou (Triangle)
Figure 7. Areal proportions for 17 LST grades
While in fact, as it is too hard to find sunny data without any
cloud in this region with such abundant water vapour, we could
only make our experiment during the season of winter
(Oct~Dec). Those results may not totally answer these climatic
puzzles, but our hypotheses are partially validated.
Chongqing, Wuhan and Nanjing those traditional “stove” cities
are all located along the Yangtze River, it is said that new
“stove” cities like Hangzhou, Shanghai and Guangzhou are
coming forth recently. On the basis of LST in this region, we
choose Wuhan (113.9~114.9E, 30.1~31.1N), Nanchang
(115.5~116.5E, 28.2~29.2N) Hangzhou (119.4~120.4E,
29.6~30.6N) and Shanghai (120.9~121.9E, 30.6~31.6N) to be
samples.
5. CONCLUSIONS
Although we do not have enough proof, objectively speaking,
most climate changes do come from our human’s behaviour,
except that caused by the Earth itself. It is wise to inspect
ourselves while complaining the inconstant weather phenomena.
And with the “Buffer” character, maybe we are repaying the
debt for our ancestors who lived hundreds or thousands years
ago, and maybe the most forceful culmination is still on its way,
though we have already undergone the barbaric climate
nowadays.
Figure 8 shows the maximum (Red), mean (Green) and
minimum (Blue) LST of these four cities from 2000 to 2007.
The old “stove” city of Wuhan (Solid) has superiority in both
mean and minimum levels, but always inferior in the maximum
level. To our surprise, Nanchang (Dash Dot) holds the throne in
the level of maximum temperature, and there is obvious
fluctuation in the level of minimum, however, Hangzhou
(Dotted) and Shanghai (Long Dashes) are similar in the three
levels, there is no evident difference between them and they
both have an overall trend with no big fluctuation.
The objective recording of the Earth information by images of
satellite remote sensing promotes our discovery on the induced
factor of human beings in the abnormal global climate and high
frequency of natural disasters, as to be one of the most active
and unstable elements in the bio-sphere that laying in the
bottom of these Earth Spheres, one small step for human beings
could make giant leap of the nature. Efforts for making
penetrate understandings of the relationship between human
being and the nature are hope to have positive contributions in
the exploration of harmonious developments of our society and
the maintenance of the world’s sustainable advancements.
Limited to the disadvantage of traditional optical remote
sensing, much precious behaviour of the climate and land cover
features could not be detected, and as a study for climate
change, only a ten-year time sequence is too short to make
definite conclusions, but we hope the above results are qualified
to be references in the research of global climate change.
Figure 8. LST of maximum, mean and minimum for Wuhan
(solid), Nanchang (dash dot), Hangzhou (dotted) and Shanghai
(long dashes), 2000~2007
538
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B8. Beijing 2008
APPENDIX
REFERENCES
References from Journals:
Gao Mao-fang, Qin Zhi-hao, XU Bin. Estimation of the basic
parameters for deriving surface temperature from MODIS data.
Arid zone research, 2007 (1), pp. 113-119.
Grades
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Yan Feng, Qin Zhihao, Li Maosong, Wang Yanjiao. On urban
heat island of Shanghai city from MODIS data. Geomatics and
information science of Wuhan University, 2007 (7), pp. 576580.
Nitin K. Singh, Sreekala G. Bajwa. Analysis of evolving
surface urban heat island in NW Arkansas. 2007 IEEE region 5
technical conference, April 20-21, Fayetteville, AR, pp. 409414.
He Xuezhao, Shi Peijun, Gong Daoyi. Spatial features of the
coupling between sprint NDVI and Temperature over northern
hemisphere. Acta Geographica Sinica, 2002 (5), pp. 505-514.
LST Range (K)
< 285
285 ~ 286
286 ~ 287
287 ~ 288
288 ~ 289
289 ~ 290
290 ~ 291
291 ~ 292
292 ~ 293
293 ~ 294
294 ~ 295
295 ~ 296
296 ~ 297
297 ~ 298
298 ~ 299
299 ~300
> 300
Appendix A. Temperature grades for part 4
539
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B8. Beijing 2008
540